CN105095305B - A kind of generation method and device of personal page - Google Patents
A kind of generation method and device of personal page Download PDFInfo
- Publication number
- CN105095305B CN105095305B CN201410213762.XA CN201410213762A CN105095305B CN 105095305 B CN105095305 B CN 105095305B CN 201410213762 A CN201410213762 A CN 201410213762A CN 105095305 B CN105095305 B CN 105095305B
- Authority
- CN
- China
- Prior art keywords
- internet application
- user
- correlation
- interest
- internet
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The present invention relates to a kind of generation methods of personal page, comprising: user behavior data of the record user in the first internet application and the second internet application；The Interests User distribution characteristics of target object in the first internet application and the second internet application is analyzed according to user behavior data；The relative coefficient between the object in the object and the second internet application in the first internet application is calculated according to Interests User distribution characteristics；First object of interest of first user in first internet application is obtained when detecting that the first user logs in the second internet application；It is greater than the second object of interest of predetermined value from the relative coefficient obtained between the first object of interest in the second internet application, and generates the personal page of the first user according to the second object of interest.The above method can solve the new user data Sparse Problems of personal page generating process.
Description
Technical field
The present invention relates to internet data process field more particularly to the generation methods and device of a kind of personal page.
Background technique
With the development of internet technology, personalized recommendation system has been widely used in online e-commerce platform.
So-called personalized recommendation is the Characteristic of Interest and buying behavior according to user, to the interested information of user recommended user and quotient
Product, so as to save the time that user searches for information or commodity.
Currently used personalized recommendation algorithm predicts the interested commodity of user by the historical behavior using user.
Therefore, there is no any behavior or considerably less behavior on this platform as some user, then existing technology is not done
The personalized page is generated for user in method.
Summary of the invention
In view of this, the embodiment of the present invention provides the generation method and device of a kind of personal page, can solve above-mentioned
When be not recorded user behavior or user behavior record data it is very few when personal page generation problem.
A kind of generation method of personal page, comprising:
Record user behavior data and root of the user in the first internet application and the second internet application
The interest of target object in first internet application and the second internet application is analyzed according to user behavior data
User distribution feature；
Object and the second interconnection in first internet application are calculated according to the Interests User distribution characteristics
The relative coefficient between object in net application program；
First user is obtained when detecting that the first user logs in second internet application described
The first object of interest in first internet application；And
It is greater than from the relative coefficient between first object of interest is obtained in second internet application
Second object of interest of predetermined value, and generate according to second object of interest personal page of first user.
A kind of generating means of personal page, comprising:
User behavior logging modle, for recording user in the first internet application and the second internet application
In user behavior data and first internet application and the second Internet application are analyzed according to user behavior data
The Interests User distribution characteristics of target object in program；
Correlation module, for being calculated in first internet application according to the Interests User distribution characteristics
The relative coefficient between object in object and the second internet application；
Module is obtained, for obtaining described the when detecting that the first user logs in second internet application
First object of interest of one user in first internet application；And
Page generation module, for being obtained between first object of interest from second internet application
Relative coefficient be greater than the second object of interest of predetermined value, and generate first user's according to second object of interest
Personal page.
According to above-mentioned scheme, when there is no user behaviors to record data in some internet application by a user
When, the data acquisition user can be recorded indirectly by the user behavior in other internet applications answer in current internet
With the object of interest in program, to solve the new user data Sparse Problems of personal page generating process.
For above and other objects, features and advantages of the invention can be clearer and more comprehensible, preferred embodiment is cited below particularly,
And cooperate institute's accompanying drawings, it is described in detail below.
Detailed description of the invention
Fig. 1 is a kind of configuration diagram of communication system Internet-based.
Fig. 2 is the structural block diagram of a server.
Fig. 3 is the flow chart of the personal page generation method of first embodiment.
Fig. 4 is the flow chart of the personal page generation method of second embodiment.
Fig. 5 is the structural block diagram of the personal page generating means of the 5th embodiment.
Fig. 6 is the structural block diagram of the personal page generating means of sixth embodiment.
Fig. 7 is that the personal page of the 7th embodiment generates the structural block diagram of system.
Fig. 8 is that the personal page of the 8th embodiment generates the structural block diagram of system.
Specific embodiment
Further to illustrate that the present invention is the technical means and efficacy realizing predetermined goal of the invention and being taken, below in conjunction with
Attached drawing and preferred embodiment, to specific embodiment, structure, feature and its effect according to the present invention, detailed description is as follows.
It refering to fig. 1, is a kind of configuration diagram of communication system Internet-based.As shown in Figure 1, this system includes
Service layer 10, transport layer 20 and terminating layer 30.
Service layer 10 refers to the server system being deployed in internet, and interior operation has for providing various internets clothes
The server-side application (hereinafter referred to as internet application) of business.The example of above-mentioned Internet service includes, but not
It is limited to, online shopping website, game on line website, social networks, Online Music website, Online Video website, online reading net
Stand etc..
Referring to Fig.2, its structural block diagram for a server of above-mentioned server system.It is appreciated that above-mentioned clothes
Business device system may include one or more servers shown in Fig. 2.As shown in Fig. 2, server includes: memory 102, processing
Device 104 and network module 106.It is appreciated that structure shown in Fig. 2 is only to illustrate, the structure of server is not caused
It limits.For example, server may also include than shown in Fig. 2 more perhaps less component or with different from shown in Fig. 2
Configuration.
Memory 102 can be used for storing software program and module, and processor 104 is stored in memory 102 by operation
Interior software program and module, thereby executing various function application and data processing.Memory 102 may include that high speed is random
Memory may also include nonvolatile memory, such as one or more magnetic storage device, flash memory or other are non-volatile
Property solid-state memory.In some instances, memory 102 can further comprise the storage remotely located relative to processor 104
Device, these remote memories can pass through network connection to server.The example of above-mentioned network includes but is not limited to internet, enterprise
Industry intranet, local area network, mobile radio communication and combinations thereof.
Transmission module 106 is for receiving and transmitting network signal.Above-mentioned network signal may include wireless signal or have
Line signal.In an example, above-mentioned network signal is cable network signal.At this point, transmission module 106 may include processor,
The elements such as random access memory, converter, crystal oscillator.
Above-mentioned software program and module includes: operating system 122, service module 124, user behavior logging modle
126, personal page generation module 128.Wherein operating system 122 may be, for example, LINUX, UNIX, WINDOWS, may include
Various component softwares and/or drive for management system task (such as memory management, storage equipment control, power management etc.)
It is dynamic, and can mutually be communicated with various hardware or component software, to provide the running environment of other software component.Service module
124 operate on the basis of operating system 122, and are monitored by the network service of operating system 122 come the request of automatic network, root
Corresponding data processing is completed according to request, and returns to processing result to client.That is, service module 124 is used for visitor
Family end provides above-mentioned Internet service.
Further, service module 124 is also responsible for mentioning user behavior data other than providing above-mentioned connection net service
Supply user behavior logging modle 126 is recorded.Above-mentioned user behavior data refers to user using the internet provided
Browse, click during service, collecting, buying, pay close attention to a certain physical item, virtual objects, service, contact person record number
According to.In short, to those skilled in the art, it is equal can arbitrarily to embody user interest, habit, the behavioral data of preference
It may include within the scope of above-mentioned user behavior data.User behavior logging modle 126 is used for the use for providing service module 124
Family behavioral data is stored in file or in database.
Personal page generation module 128 is used to that the personalized page to be generated for user according to user behavior data.For example,
Generate meet the commodity of customer consumption interest, service, contact person the recommendation page.Personal page generation module 128 it is specific
Implementation process will be described as after in conjunction with specific embodiments.
Above-mentioned module refers to the software program module being stored in memory 102, when it is executed by processor 104,
Just it is able to achieve function desired by each module.However, above-mentioned module can also use hardware, for example, the integrated electricity of special applications
Road (Application Specific Integrated Circuit, ASIC), field programmable gate array (Field
Programmable Gate Array, FPGA) etc. substitute.
Again referring to Fig. 1, in the present embodiment, service layer 10 includes first server system 11 and second server system
12, it is inside deployed with the first internet application and the second internet application respectively.Although in addition, being shown in Fig. 1
It is deployed in different server systems respectively, however the first internet application and the second internet application can also be with
It is deployed in same server system.
First internet application and the second internet application can provide the Internet service of same type, can also
To provide different types of connection net service respectively.In one embodiment, the first internet application and the second interconnection
Net application program is provided which online shopping service, but it is nonoverlapping that the commodity of its sale, service, which at least have part,.At one
In embodiment, the first internet application and the second internet application are provided which online shopping service, but it is sold
Commodity, service be completely it is nonoverlapping.In one embodiment, the first internet application provides online shopping clothes
Business, and the second internet application provides online music and plays service.In one embodiment, the first Internet application journey
Sequence provides online music and plays service, and the second internet application provides video playing service on line.
Transport layer 20 refers to the network of various levels, such as backbone network, Metropolitan Area Network (MAN), local area network, mobile radio communication, wireless office
Domain net etc., transport layer 20 are responsible for linking Internet and data transmission between service layer 10 and terminating layer 30.In general, it services
Layer 10 server system be generally deployed in the computer room or data center of Internet Service Provider, generally can directly with
Backbone network is connected.And the various terminals in terminating layer 30 can be interconnected by accesses such as local area network, WLAN, mobile radio communications
Net, so as to the intercommunication with server 10.
The specific example of terminal in terminating layer 30 includes but is not limited to desktop computer, laptop, tablet computer
(the including but not limited to iPad of Apple Inc. and other operation Apple Inc.'s IOS operating system touch-screen equipment, operation it is micro-
The touch-screen equipment of soft company's Windows operating system and the touch-screen equipment for running Google's Android operation system), it is mobile
Phone, smart phone (including but not limited to Apple Inc.'s IOS operating system, Windows Mobile operating system, black mould operation
System, the smart phone for filling in spot operating system), E-book reader, any aforementioned operation system of operation or other operation systems
System car-mounted terminal and wearable electronic terminal (including but not limited to intelligent glasses, wrist-watch, bracelet, wrist strap and other
Ornaments).
As shown in Figure 1, desktop computer 31, which is connect by coaxial cable with router 301, accesses internet, notebook electricity
Brain 32 and smart phone 33 are by the access of accessing wirelessly hot spot 302 internet, and tablet computer 34 passes through with wearable device 35
Internet is accessed in mobile base station 303.It is appreciated that no matter which kind of mode is used to access internet, the various ends in terminating layer 30
The various Internet services that the logical access server 10 in end provides.
Various terminals in terminating layer 30 are based on browser or local application (Native Application) mentions
It is interacted for the front-end interface of above-mentioned various Internet services, and by these interfaces with user, user can be at these
The behaviour such as above-mentioned browsing, click, collection, purchase, a certain physical item of concern, virtual objects, service, contact person are completed in interface
Make.After the input operation for detecting user, corresponding data can be sent to corresponding internet by transport layer 20 by terminal
Application program.And internet application in addition to the normal data processing of progress and returns the result after receiving these data
Outside, the behavioral data of also recordable user.
Refering to Fig. 3, for the flow chart for the personal page generation method that first embodiment provides.As shown in figure 3, above-mentioned
Method the following steps are included:
Step S101, user behavior of the record user in the first internet application and the second internet application
Data.
First internet application and the second internet application can provide the Internet service of same type, can also
To provide different types of connection net service respectively.In one embodiment, the first internet application and the second interconnection
Net application program is provided which online shopping service, but it is nonoverlapping that the commodity of its sale, service, which at least have part,.At one
In embodiment, the first internet application and the second internet application are provided which online shopping service, but it is sold
Commodity, service be completely it is nonoverlapping.In one embodiment, the first internet application provides online shopping clothes
Business, and the second internet application provides online music and plays service.In one embodiment, the first Internet application journey
Sequence provides online music and plays service, and the second internet application provides video playing service on line.
Above-mentioned user behavior, which refers to, completes browsing, point in the first internet application and the second internet application
It hits, collect, buying, paying close attention to a certain physical item, virtual objects, service, the operation of contact person, these operations generate data and are
Above-mentioned user behavior data.
Step S102, first internet application and the second Internet application journey are analyzed according to user behavior data
The Interests User distribution characteristics of target object in sequence.
Above-mentioned target object for example refer to article involved in internet application, commodity, virtual objects, stage property,
The objects such as service, contact person, the object are that internet answers program to provide the basic unit of Internet service.Interests User distribution is special
Sign refers to distribution characteristics of the user to the subject interests in all users.Whether one user is interested in some object
It can be defined by those of ordinary skill in the art according to specific demand and specific application scenarios.For example, for
Line shopping website, can define browsing, collection or buy the user of some commodity is that the commodity are interested.For social network
Network can define the Interests User for paying close attention to the listener of some contact person as the contact person.For the Interests User of target object
Distribution characteristics, those of ordinary skill in the art can be described using any appropriate data structure.
In one embodiment, above-mentioned Interests User distribution characteristics can be characterized using user characteristics vector.?
That is step S101 can comprise the following steps that
N number of user of the first internet application or the second internet application is obtained, wherein N is natural number；With
And
Whether analysis target object is the object of interest of N number of user to form the user characteristics vector of the object
[a1, a2 ... aN], wherein the value of an (1≤n < N) characterizes user n to the interest-degree of the object.
For example, having 5 users (1,2,3,4,5) in an online shopping site, for commodity A, user 1,3,5 is purchased
It bought.Then the user characteristics vector of commodity A can be expressed as [1,0,1,0,1].As can be seen that in this example, use 1 is indicated pair
The commodity are interested, and use 0 indicates to lose interest in the commodity.It is appreciated that 1 can be replaced with 0 value with any other value.
In the expression formula of above-mentioned user characteristics vector, 0 indicates to lose interest in completely, and 1 indicates completely interested, remaining
Value from 0 to 1, indicate interest-degree it is higher and higher.For example, in above-mentioned example, in conjunction with user interest degree user characteristics to
Amount can be expressed as [1,0,0.5,0,0.7].And specific interest-degree information can be obtained according to the different behavioural analysis of user.
For example, buying a certain commodity for user, then user is 1 to the interest-degree of the commodity, and browses the related pages of the commodity, emerging
Interesting degree may be defined as 0.5, and collect the commodity, then interest-degree can be defined as 0.8.
Step S103, the target pair in first internet application is calculated according to the Interests User distribution characteristics
As the relative coefficient between the target object in the second internet application.
It is appreciated that the first internet application and the second internet application can have the identical user in part,
The user of this part phase all has object of interest in the first internet application and the second internet application.Namely
It says, same user account number can log in different internet applications simultaneously.Alternatively, two different are bound between each other
The user account number of first internet application and the second internet application can also be considered as the account number of same user.
User identical for this part, the target in the first internet application and the second internet application
Object can form corresponding user characteristics vector, then can be calculated between two target objects according to two vectors
Relative coefficient.For example, cosine similarity or Pearson correlation coefficient can be calculated as upper for two given vectors
The relative coefficient stated.
Step S104, described first is obtained when detecting that the first user logs in second internet application to use
First object of interest of the family in first internet application.
For the user for not yet using the second internet application, in the second internet application simultaneously
There is no user behaviors to record data, obtains therefore, it is impossible to directly answer the user behavior of program to record data according to the second internet
Take object of interest of first user in the second internet application.In the present embodiment, in which case, the is obtained first
First object of interest of one user in the first internet application.First object of interest for example refers to above-mentioned commodity, clothes
Business, contact person etc..First object of interest may include one or more.
Step S105, it is obtained and the correlation between first object of interest from second internet application
Coefficient is greater than the second object of interest of predetermined value, and the personalized page of first user is generated according to second object of interest
Face.
As described above, having obtained the first internet application and the second internet application in step s 103
Relative coefficient between object, therefore in step s105, all phase relations between the first object of interest can be obtained
Coefficient is greater than the second object of interest of predetermined value.Second object of interest for example refers to above-mentioned commodity, service, contact person etc..
Second object of interest may include one or more.
It can be that the first user generates the personalized page according to the second object of interest after obtaining the second object of interest.
For example, can be used for buying the commodity according to commodity generation or be added to the link of shopping cart.Being generated according to contact person can be used for
It pays close attention to the contact person's or adds the link that the contact person is good friend.After generating these personalized pages, it can be returned
Back to user terminal to be shown.
According to the method for the present embodiment, when there is no user behaviors to record in some internet application by a user
When data, the data acquisition user can be recorded indirectly by the user behavior in other internet applications and interconnected currently
Object of interest in net application program, to solve the new user data Sparse Problems of personal page generating process.
Second embodiment provides a kind of personal page generation method, difference similar to method shown in Fig. 3
Be, step S103 the following steps are included:
Step S201, according in the shared user group of first internet application and the second internet application
All users and corresponding all object of interest calculate correlation matrix R '.
Three groups can be divided into, wherein using for all users of arbitrary two internet applications refering to following table
Family group 1 indicates only there is object of interest in the first internet application, and user group 2 (shared user group) is in the first internet
Object of interest is all had in application program and the second internet application, and user group 3 is only in the second internet application
In have object of interest.
The user grouping signal of the different internet applications of table 1
It is whether interested according to the user in user group 2 for object a1, a2, b1, b2, b3 according to upper table, it can distinguish
Form user characteristics vector [1,1], [0,1], [1,1], [1,0], [1,0].By features described above vector, it can be seen that object a1 with
Relative coefficient is higher between object b1, and relative coefficient is lower between object a1 and object b2 and b3.According to any two
The user characteristics vector of object, can calculate the relative coefficient between two objects, can form correlation matrix R ' into face,
And element R ' ij in R ' is indicated in the first internet application in object i and the second internet application between object j
Relative coefficient.Therefore, according to correlation matrix R ', the first internet application and the second Internet application can be obtained
Relative coefficient in program between any two object.
Step S202, according to the first internet of user distribution feature calculation of object in first internet application
Correlation matrix S in application program between object.
Table 1 is please referred to again, for the object a1 and a2 in the first internet application, according to user group 1 and user group 2
Interior all users can also form user characteristics vector, such as the user characteristics vector of object a1 is [1,1,1,1], object a2
User characteristics vector be [1,1,0,1].It can be seen that from user characteristics vector with higher between object a1 and object a2
Correlation.In step S202, any two object is calculated according to the user characteristics vector of object in the first internet application
Between correlation and form correlation matrix S, elements Si i ' in correlation matrix S is indicated before object i and object i '
Correlation.Therefore, according to correlation matrix S, the phase in the first internet application between any two object can be obtained
Guan Xing.
Step S203, according to the second internet of user distribution feature calculation of object in second internet application
Correlation matrix S ' in application program between object.
It is similar to correlation matrix S, it can be according to all users in user group 2 and user group 3 whether to the second internet
Correlation matrix S ' between object in application program.Element S in correlation matrix S ' ' jj ' indicates object j and object j '
Correlation before.Therefore, according to correlation matrix S；, can obtain in the second internet application any two object it
Between correlation.
Step S204, according to the correlation matrix R ', S, S ' calculate the first internet application and the second internet
Relative coefficient in application program between any object.
In one embodiment, step S204 is the following steps are included: calculate the object m in the first internet application
Relative coefficient Rmj=Smi × R ' ij between the object j in the second internet application, wherein R ' ij is first mutual
The correlation between the object i and object j in the second internet application in networked application programs, Smi are the first interconnection
Correlation in net application program between object m and object i.
Table 1 is please referred to again, if only forming user characteristics vector according to user group 2, and object a2 is answered with the second internet
Correlation with any object in program is lower, that is to say, that when the object of interest of the first user above-mentioned only has object
When a2, corresponding correlation can not be obtained higher than in the second internet application of certain value according to correlation matrix R '
Object.And according to the present embodiment, the relative coefficient between object a2 and object b1 can be defined as object a2 and object a1
The relative coefficient of relative coefficient before × between object a1 and object b1.Since object a1 and a2 is highly relevant and right
As a1 and object b1 are highly relevant, therefore, object a2 is associated with object b1 indirectly by object a1.
It is appreciated that wherein any one object can pass through for L object in the first internet application
Object in other (L-1) a object and the second internet application be between it is interrelated.That is, formula Rmj
The possible calculated result of=Smi × R ' ij has L-1, for this L-1 calculated result, can be maximized, can also be to L-1
A calculated result calculates average value as final relative coefficient.
In one embodiment, step S204 is the following steps are included: calculate the first internet application and second mutually
Relative coefficient Rin=R ' ij × S ' jn in networked application programs between any object, wherein R ' ij is that the first internet is answered
With the correlation between the object i and object j in the second internet application in program, S ' jn is the second Internet application
Correlation in program between object j and object n.
It is appreciated that wherein any one object can pass through for M object in the second internet application
Object in other M-1 objects and the first internet application be between it is interrelated.That is, formula Rmj=
The possible calculated result of Smi × R ' ij has M-1, for this M-1 calculated result, can be maximized, can also be to L-1
Calculated result calculates average value as final relative coefficient.
In above mode, the correlation between object i and object j can be obtained by an indirect object (m or n)
It takes, therefore, aforesaid way can be summarized as level-one indirect correlation, and need when the calculating of level-one indirect correlation still can not obtain
When the result wanted, it is also contemplated that calculated using multistage indirect correlation.
In one embodiment, step S204 is the following steps are included: calculate the first internet application and second mutually
Relative coefficient Rmn=Smi × R ' ij × S ' jn in networked application programs between any object, wherein R ' ij is the first interconnection
The correlation between the object i and object j in the second internet application in net application program, Smi are the first internet
Correlation in application program between object m and object i, S ' jn be the second internet application between object j and object n
Correlation.
It is appreciated that there is M in the second internet application when having L object in the first internet application
When a object, the calculated result of formula Rmn=Smi × R ' ij × S ' jn is planted with (L-1) * (M-1), for this (L-1) * (M-
1) a calculated result can be maximized, and can also calculate a calculated result of (L-1) * (M-1) average value as final phase
Close property coefficient.
In addition, essentially by other objects as medium in step S204, realize the first internet should have program with
The calculating of object dependencies between second internet application.As such, it can be that only when can not according to correlation matrix R '
Just make when obtaining the correlation in the first internet application between object i and the object j in the second internet application
With indirect calculation.That is, when element corresponding with commodity m, Huo Zhesuo are not present in the correlation matrix R '
Have the value of element corresponding with commodity m just calculated when being respectively less than the predetermined value relative coefficient Rmn=Smi × R ' ij ×
S'jn.Use other formula calculate when can with and so on.
According to the method for the present embodiment, when there is no user behaviors to record in some internet application by a user
When data, the data acquisition user can be recorded indirectly by the user behavior in other internet applications and interconnected currently
Object of interest in net application program, to solve the new user data Sparse Problems of personal page generating process.
3rd embodiment provides a kind of generation method of personal page.In the present embodiment, the first internet application
For online shopping site A, the second internet application is online shopping site B.The wherein commodity that online shopping site A is sold
It mainly include general merchandise, and the staple that online shopping site B is sold is books.Online shopping site A and online shopping
Website B receives third party's the Internet account and logs in.
For an account number user a of third party's the Internet account, often do shopping in the A of website, but do not step on
Website B is crossed in land.Therefore, after user a logs in website B, due to not having historical record data, website B can not generate a for it
The page of property.At this point, by setting up a web site the commodity and website B of A sale in the user that website A and website B were logged in simultaneously
Correlation matrix between the commodity of sale, and user a is obtained in the A of website between object of interest according to correlation matrix
Relative coefficient is greater than the object of predetermined value as target object, and is that user a generates personalized page according to target object
Face.
According to the method for the present embodiment, behavior record of one user in the A of website can be mapped by correlation matrix
Into the commodity of website B, to solve the problems, such as that the personal page of new user in the B of website generates.
Fourth embodiment provides a kind of generation method of personal page.In the present embodiment, the first internet application
For social networks (Social Network System, SNS) website C, the second internet application is online shopping site D.
Wherein website C provides social networking service, user can wherein create, share personal, tissue social networks and with social activity
Contact person in network communicates, and the staple that online shopping site D is sold is books.Online shopping site D receives
The user account number of website C logs in.
It is very active in the C of website for the user account number user b of website C, but and not landed website D excessively.
Therefore, after user b logs in website D, due to not having historical record data, website D can not generate personalized page for it
Face.And in social networks, a contact person, which is concerned, listens to or establishes friend relation substantially, can analogize to a commodity
Purchased or collection.That is, can equally be formed according to above-mentioned method for each contact person in social networks
User characteristics vector, and further calculating is compared to obtain the connection in the C of website with the user characteristics vector of books in the D of website
It is the relative coefficient between the books in people and website D.In this way, after user b logs in website D, it is available its in website
The contact person or good friend of concern in C, according to correlation matrix acquisition associate between people relative coefficient be greater than it is predetermined
The set of the books of value, and user b is further combined into according to the collection of the books of acquisition and generates the personalized page.
According to the method for the present embodiment, the net of different type Internet service can will be provided by two by correlation matrix
Association is established between standing, to solve the problems, such as that the personal page of new user in the D of website generates.
Refering to Fig. 5, for the structural block diagram for the personal page generating means that the 5th embodiment provides.As shown in figure 5, this
The device of embodiment includes: user behavior logging modle 51, feature acquisition module 52, correlation module 53, object acquisition module
54 and page generation module 55.
User behavior logging modle 51 is for recording user in the first internet application and the second Internet application journey
User behavior data in sequence.
Feature obtains module 52 and is used to analyze first internet application and second mutually according to user behavior data
The Interests User distribution characteristics of target object in networked application programs.
Correlation module 53 is used to be calculated in first internet application according to the Interests User distribution characteristics
Object and the second internet application in object between relative coefficient.
Object acquisition module 54 is used for the acquisition when detecting that the first user logs in second internet application
First object of interest of first user in first internet application.
Page generation module 55 be used for from second internet application obtain with first object of interest it
Between relative coefficient be greater than the second object of interest of predetermined value, and first user is generated according to second object of interest
Personal page.
The other details of device about the present embodiment can also further regard to the associated description of first embodiment.
According to the device of the present embodiment, when there is no user behaviors to record in some internet application by a user
When data, the data acquisition user can be recorded indirectly by the user behavior in other internet applications and interconnected currently
Object of interest in net application program, to solve the new user data Sparse Problems of personal page generating process.
Refering to Fig. 6, for the structural block diagram for the personal page generating means that sixth embodiment provides.As shown in fig. 6, this
The device of embodiment and the device of Fig. 5 are similar, the difference is that, correlation module 53 includes: the first computing module 61,
Two computing modules 62, third computing module 63 and correlation calculations module 64.
First computing module 61 is used for being total to according to first internet application and the second internet application
It enjoys all users and corresponding all object of interest in user group and calculates correlation matrix R '.
Second computing module 62 is used for the user distribution feature calculation according to object in first internet application
Correlation matrix S in first internet application between object.
Third computing module 63 is used for the user distribution feature calculation according to object in second internet application
Correlation matrix S ' in second internet application between object.
Correlation calculations module 64 be used for according to the correlation matrix R ', S, S ' calculate the first internet application with
Relative coefficient in second internet application between any object.
The other details of device about the present embodiment can also further regard to the associated description of second embodiment.
According to the device of the present embodiment, when there is no user behaviors to record in some internet application by a user
When data, the data acquisition user can be recorded indirectly by the user behavior in other internet applications and interconnected currently
Object of interest in net application program, to solve the new user data Sparse Problems of personal page generating process.
Refering to Fig. 7, the structural block diagram of system is generated for the personal page that the 7th embodiment provides.On as shown in fig. 7,
The system stated includes: first server 71 and second server 72.
First server 71 includes the first user behavior logging modle 711 and the first analysis module 712, second server 72
Including the second logging modle 721 and the second logging modle 722.Wherein, the first logging modle 711 is mutual first for recording user
User behavior data in networked application programs, the first analysis module 712 are used for the data recorded according to the first logging modle 711
Analyze the Interests User distribution characteristics of target object in the first internet application.Second logging modle 721 is used for recording
User behavior data of the family in the second internet application, the second analysis module 722 are used for according to the second logging modle 721
The data of record analyze the Interests User distribution characteristics of target object in the second internet application.
Second server 72 further includes correlation module 723, object acquisition module 724 and page generation module 725.
Correlation module 723 is used to be calculated in first internet application according to the Interests User distribution characteristics
Target object and the second internet application in target object between relative coefficient.
Object acquisition module 724 is used for the acquisition when detecting that the first user logs in second internet application
First object of interest of first user in first internet application.
Page generation module 725 be used for from second internet application obtain with first object of interest it
Between relative coefficient be greater than the second object of interest of predetermined value, and first user is generated according to second object of interest
Personal page.
The other details of device about the present embodiment can also further regard to the associated description of first embodiment.
According to the system of the present embodiment, when there is no user behaviors to record in some internet application by a user
When data, the data acquisition user can be recorded indirectly by the user behavior in other internet applications and interconnected currently
Object of interest in net application program, to solve the new user data Sparse Problems of personal page generating process.
Sixth embodiment provides a kind of personal page generation system, difference similar to system shown in fig. 5
It is, correlation module 723 includes: the first computing module 61, the second computing module 62, third computing module 63 and correlation
Computing module 64.
First computing module 61 is used for being total to according to first internet application and the second internet application
It enjoys all users and corresponding all object of interest in user group and calculates correlation matrix R '.
Second computing module 62 is used for the user distribution feature calculation according to object in first internet application
Correlation matrix S in first internet application between object.
Third computing module 63 is used for the user distribution feature calculation according to object in second internet application
Correlation matrix S ' in second internet application between object.
Correlation calculations module 64 be used for according to the correlation matrix R ', S, S ' calculate the first internet application with
Relative coefficient in second internet application between any object.
The other details of device about the present embodiment can also further regard to the associated description of second embodiment.
According to the system of the present embodiment, when there is no user behaviors to record in some internet application by a user
When data, the data acquisition user can be recorded indirectly by the user behavior in other internet applications and interconnected currently
Object of interest in net application program, to solve the new user data Sparse Problems of personal page generating process.
The above described is only a preferred embodiment of the present invention, be not intended to limit the present invention in any form, though
So the present invention is disclosed as above with preferred embodiment, and however, it is not intended to limit the invention, anyone skilled in the art, not
It is detached within the scope of technical solution of the present invention, when the technology contents using the disclosure above are modified or are modified to equivalent change
The equivalent embodiment of change, but without departing from the technical solutions of the present invention, implement according to the technical essence of the invention to above
Any brief introduction modification, equivalent variations and modification made by example, all of which are still within the scope of the technical scheme of the invention.
Claims (20)
1. a kind of generation method of personal page, which is characterized in that the described method includes:
Record user behavior data of the user in the first Internet application and the second Internet application；
According to the user behavior data analyze first Internet application in the second Internet application target object it is emerging
Interesting user distribution feature；
It is calculated according to the Interests User distribution characteristics in the object and the second Internet application in first Internet application
Object between relative coefficient, comprising:
According to first Internet application and all users in the shared user group of the second Internet application and corresponding institute
Interesting calculation and object correlation matrix R ', wherein each user in the shared user group is in first Internet application
Object of interest is all had in the second Internet application；
It is calculated in the first Internet application and the second Internet application between any object according at least to the correlation matrix R '
Relative coefficient；
First user is obtained when detecting that the first user logs in second Internet application in first interconnection
The first object of interest in net application；And
It is greater than predetermined value from the relative coefficient obtained between first object of interest in second Internet application
Second object of interest, and generate according to second object of interest personal page of first user.
2. the method as described in claim 1, which is characterized in that first Internet application according to user behavior analysis
Interests User distribution characteristics with object in the second Internet application includes:
N number of user of first Internet application or the second Internet application is obtained, wherein N is natural number；And
Analyze whether the object is the object of interest of N number of user to form the user characteristics vector [a of the object1,
a2,…aN], wherein anThe value of (1≤n < N) characterizes in N number of user nth user to the interest-degree of the object.
3. method according to claim 2, which is characterized in that described to calculate described the according to the Interests User distribution characteristics
The relative coefficient between the object in object and the second Internet application in one Internet application includes:
The cosine similarity or Pearson correlation coefficient between different objects are calculated according to the user characteristics vector.
4. method according to claim 2, which is characterized in that described to obtain first Internet application or the second internet
Application N number of user include:
It obtains simultaneously in first Internet application and the second Internet application with N number of user of object of interest.
5. method according to any of claims 1-4, which is characterized in that described according to the Interests User distribution characteristics meter
The relative coefficient calculated between the object in object and the second Internet application in first Internet application further wraps
It includes:
According between object in the first Internet application of user distribution feature calculation of object in first Internet application
Correlation matrix S；
According between object in the second Internet application of user distribution feature calculation of object in second Internet application
Correlation matrix S '；And
According to the correlation matrix R ', S, S ' it calculates in the first Internet application and the second Internet application between any object
Relative coefficient.
6. method as claimed in claim 5, which is characterized in that described according to the correlation matrix R ', S, S ' calculate first
Relative coefficient in Internet application and the second Internet application between any object includes:
Calculate the relative coefficient R in the first Internet application and the second Internet application between any objectmn=Smi×R’ij
×S’jn, wherein R 'ijFor the correlation between the object i and object j in the second Internet application in the first Internet application,
SmiFor the correlation in the first Internet application between object m and object i, S 'jnFor object j in the second Internet application with it is right
As the correlation between n.
7. method as claimed in claim 6, which is characterized in that when in the R ' be not present element corresponding with object m, or
The value of all elements corresponding with object m of person just calculates the relative coefficient R when being respectively less than the predetermined valuemn=Smi×R’ij
×S’jn。
8. method as claimed in claim 5, which is characterized in that described according to the correlation matrix R ', S, S ' calculate first
Relative coefficient in Internet application and the second Internet application between any object includes:
Calculate the relative coefficient R in the first Internet application and the second Internet application between any objectmj=Smi×R’ij,
Wherein R 'ijFor the correlation between the object i and object j in the second Internet application in the first Internet application, SmiIt is
Correlation in one Internet application between object m and object i.
9. method as claimed in claim 5, which is characterized in that described according to the correlation matrix R ', S, S ' calculate first
Relative coefficient in Internet application and the second Internet application between any object includes:
Calculate the relative coefficient R in the first Internet application and the second Internet application between any objectin=R 'ij×S
’jn, wherein R 'ijFor the correlation between the object i and object j in the second Internet application in the first Internet application, S 'jn
For the correlation in the second Internet application between object j and object n.
10. the method as described in claim 1, which is characterized in that described to generate described first according to second object of interest
The personal page of user includes: to generate the link for being directed toward second object of interest.
11. a kind of generating means of personal page, which is characterized in that described device includes:
User behavior logging modle, for recording user behavior of the user in the first Internet application and the second Internet application
Data；
Feature obtains module, for being analyzed in first Internet application and the second Internet application according to user behavior data
The Interests User distribution characteristics of target object；
Correlation module, for calculating the object and in first Internet application according to the Interests User distribution characteristics
The relative coefficient between object in two Internet applications is specifically used for according to first Internet application and the second interconnection
All users and corresponding all object of interest calculate correlation matrix R ' in the shared user group of net application, wherein described total
The each user enjoyed in user group all has object of interest in first Internet application and the second Internet application；At least
The correlation between the first Internet application and any object in the second Internet application is calculated according to the correlation matrix R '
Coefficient；
Object acquisition module is used for obtaining described first when detecting that the first user logs in second Internet application
First object of interest of the family in first Internet application；And
Page generation module, for being obtained and the correlation between first object of interest from second Internet application
Coefficient is greater than the second object of interest of predetermined value, and the personalized page of first user is generated according to second object of interest
Face.
12. device as claimed in claim 11, which is characterized in that the user behavior logging modle is according to user behavior analysis
First Internet application and the Interests User distribution characteristics of object in the second Internet application include:
N number of user of first Internet application or the second Internet application is obtained, wherein N is natural number；And
Analyze the object whether be N number of user object of interest to formed the object user characteristics vector [a1,
A2 ... aN], wherein the value of an (1≤n < N) characterizes in N number of user nth user to the interest-degree of the object.
13. device as claimed in claim 12, which is characterized in that the correlation module is used for:
The cosine similarity or Pearson correlation coefficient between different objects are calculated according to the user characteristics vector.
14. device as claimed in claim 12, which is characterized in that the user behavior logging modle obtains first interconnection
Net is applied or N number of user of the second Internet application includes:
It obtains simultaneously in first Internet application and the second Internet application with N number of user of object of interest.
15. such as the described in any item devices of claim 11-14, which is characterized in that the correlation module is further used for:
According between object in the first Internet application of user distribution feature calculation of object in first Internet application
Correlation matrix S；
According between object in the second Internet application of user distribution feature calculation of object in second Internet application
Correlation matrix S '；And
According to the correlation matrix R ', S, S ' it calculates in the first Internet application and the second Internet application between any object
Relative coefficient.
16. device as claimed in claim 15, which is characterized in that the correlation module according to the correlation matrix R ',
S, S ' calculates the relative coefficient in the first Internet application and the second Internet application between any object and includes:
Calculate relative coefficient Rmn=Smi × R ' in the first Internet application and the second Internet application between any object
Ij × S ' jn, wherein R ' ij be the first Internet application in object i and the object j in the second Internet application between it is related
Property, Smi is the correlation in the first Internet application between object m and object i, and S ' jn is object j in the second Internet application
With the correlation between object n.
17. device as claimed in claim 16, which is characterized in that when in the R ' be not present element corresponding with object m,
Or the value of all elements corresponding with object m when being respectively less than the predetermined value correlation module just calculate the correlation
Coefficients R mn=Smi × R ' ij × S ' jn.
18. device as claimed in claim 15, which is characterized in that the correlation module according to the correlation matrix R ',
S, S ' calculates the relative coefficient in the first Internet application and the second Internet application between any object and includes:
Calculate relative coefficient Rmj=Smi × R ' in the first Internet application and the second Internet application between any object
Ij, wherein R ' ij is the correlation between the object i and object j in the second Internet application in the first Internet application, Smi
For the correlation in the first Internet application between object m and object i.
19. device as claimed in claim 15, which is characterized in that the correlation module according to the correlation matrix R ',
S, S ' calculates the relative coefficient in the first Internet application and the second Internet application between any object and includes:
Calculate relative coefficient Rin=R ' ij × S ' in the first Internet application and the second Internet application between any object
Jn, wherein R ' ij is the correlation between the object i and object j in the second Internet application in the first Internet application, S '
Jn is the correlation in the second Internet application between object j and object n.
20. device as claimed in claim 11, which is characterized in that the page generation module is according to second object of interest
The personal page for generating first user includes: to generate the link for being directed toward second object of interest.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410213762.XA CN105095305B (en) | 2014-05-20 | 2014-05-20 | A kind of generation method and device of personal page |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410213762.XA CN105095305B (en) | 2014-05-20 | 2014-05-20 | A kind of generation method and device of personal page |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105095305A CN105095305A (en) | 2015-11-25 |
| CN105095305B true CN105095305B (en) | 2019-03-08 |
Family
ID=54575755
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410213762.XA Active CN105095305B (en) | 2014-05-20 | 2014-05-20 | A kind of generation method and device of personal page |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105095305B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109902252A (en) * | 2019-01-17 | 2019-06-18 | 平安城市建设科技(深圳)有限公司 | Management method, device, equipment and the storage medium of user's growth system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102289509A (en) * | 2011-08-31 | 2011-12-21 | 南京大学 | Method for obtaining, migrating and using website data |
| CN102880501A (en) * | 2012-07-24 | 2013-01-16 | 北京奇虎科技有限公司 | Realizing method, device and system for recommending applications |
| CN102999588A (en) * | 2012-11-15 | 2013-03-27 | 广州华多网络科技有限公司 | Method and system for recommending multimedia applications |
| CN103136435A (en) * | 2011-11-30 | 2013-06-05 | 深圳市快播科技有限公司 | System, method and game platform capable of recommending games in personalization mode |
-
2014
- 2014-05-20 CN CN201410213762.XA patent/CN105095305B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102289509A (en) * | 2011-08-31 | 2011-12-21 | 南京大学 | Method for obtaining, migrating and using website data |
| CN103136435A (en) * | 2011-11-30 | 2013-06-05 | 深圳市快播科技有限公司 | System, method and game platform capable of recommending games in personalization mode |
| CN102880501A (en) * | 2012-07-24 | 2013-01-16 | 北京奇虎科技有限公司 | Realizing method, device and system for recommending applications |
| CN102999588A (en) * | 2012-11-15 | 2013-03-27 | 广州华多网络科技有限公司 | Method and system for recommending multimedia applications |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105095305A (en) | 2015-11-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hew et al. | Predicting drivers of mobile entertainment adoption: a two-stage SEM-artificial-neural-network analysis | |
| CN108416625A (en) | The recommendation method and apparatus of marketing product | |
| TWI506577B (en) | System and method for identifying reviewers with incentives | |
| US20160162974A1 (en) | Personalized recommendation method and system, and computer-readable record medium | |
| CN104954410A (en) | Message pushing method, device thereof and server | |
| US20120310731A1 (en) | Method and system for displaying related product information | |
| CN102411596A (en) | Information recommendation method and system | |
| CN106251168A (en) | Information-pushing method and system | |
| CN110335123B (en) | Commodity recommendation method, system, computer readable medium and device based on social e-commerce platform | |
| CN107862352A (en) | A kind of Quick Response Code sharing method and Quick Response Code share equipment | |
| TWI579787B (en) | Systems and methods for instant e-coupon distribution | |
| CN106056408A (en) | Method and device for tracking promoter | |
| US20160196579A1 (en) | Dynamic deep links based on user activity of a particular user | |
| CN108810047A (en) | For determining that information pushes the method, apparatus and server of accuracy rate | |
| CN110765395A (en) | Method and equipment for providing novel information | |
| CN104992348A (en) | Method and device for displaying information | |
| CN110473042B (en) | Method and device for acquiring information | |
| CN109754272A (en) | The charging method and system of the web advertisement | |
| CN108399551A (en) | A kind of method and system of determining user tag and pushed information | |
| CN105027155A (en) | Unifying cloud services for online sharing | |
| CN111340522B (en) | Resource recommendation method, device, server and storage medium | |
| CN111008327B (en) | Method and device for pushing books in reading application | |
| CN105095305B (en) | A kind of generation method and device of personal page | |
| CN109639559A (en) | A kind of wechat H5 propagates method for monitoring and analyzing and relevant device | |
| US20160253332A1 (en) | Ranked navigation element |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |